VU Research Portal

VU Research Portal

Six-year healthcare trajectories of adults with anxiety and depressive disorders

Kooistra, L. C.; Wiersma, J. E.; Ruwaard, J. J.; Riper, H.; Penninx, B. W.J.H.; van Oppen,

P.

published in

Journal of Affective Disorders

2018

DOI (link to publisher)

10.1016/j.jad.2018.07.072

document version

Publisher's PDF, also known as Version of record

document license

Article 25fa Dutch Copyright Act

Link to publication in VU Research Portal

citation for published version (APA)

Kooistra, L. C., Wiersma, J. E., Ruwaard, J. J., Riper, H., Penninx, B. W. J. H., & van Oppen, P. (2018). Six-year

healthcare trajectories of adults with anxiety and depressive disorders: Determinants of transition to specialised

mental healthcare. Journal of Affective Disorders, 241, 226-234. https://doi.org/10.1016/j.jad.2018.07.072

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal ?

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address:

vuresearchportal.ub@vu.nl

Contents lists available atScienceDirect

Journal of A

ffective Disorders

journal homepage:www.elsevier.com/locate/jad

Research paper

Six-year healthcare trajectories of adults with anxiety and depressive

disorders: Determinants of transition to specialised mental healthcare

L.C. Kooistra

, J.E. Wiersma

, J.J. Ruwaard

, H. Riper

, B.W.J.H. Penninx

, P. van Oppen

institute, van der Boechorststraat 1, 1081 BT Amsterdam, The Netherlands

b_{Department of Psychiatry and the Amsterdam Public Health research institute, GGZ inGeest/VU University Medical Centre, Oldenaller 1, 1081 MJ Amsterdam, The}

Netherlands

c_{Centre for Telepsychiatry in the Region of Southern Denmark and the Clinical Institute at the University of Southern Denmark, Sdr. Boulevard 29, DK-5000 Odense,}

Denmark

A R T I C L E I N F O

Keywords: Depression Anxiety

Specialised mental healthcare Determinants

A B S T R A C T

Background: To investigate potential facilitators and barriers for patients receiving specialised mental healthcare using a longitudinal design.

Methods: Longitudinal data on 701 adult participants with a depressive and/or anxiety disorder were derived from the Netherlands Study of Depression and Anxiety (NESDA). Demographic, clinical and treatment de-terminants at baseline were assessed with self-report questionnaires and the Composite International Diagnostic Interview (CIDI 2.1). Transition to specialised mental healthcare was assessed at one, two, four, and six-year follow-up with a self-report resource use questionnaire (TiC-P).

Results: 28.3% of patients with a depressive and/or anxiety disorder transitioned from receiving no care or primary mental healthcare to specialised mental health services during six-year follow-up. The multivariate Cox regression model identified suicidal ideation, younger age, higher education level, openness to experience, pharmacological treatment, prior treatment in primary mental healthcare and perceived unmet need for help as determinants of transition, explaining 8–18% of variance.

Limitations: This study focused on baseline determinants of future transition to specialised mental healthcare. Recovery and remittance of depression and anxiety in relation to transition were not studied.

Conclusions: Not all key clinical guideline characteristics such as severity of symptoms and comorbidity were predictive of a transition to specialised mental healthcare, while non-clinical factors, such as age and perceived unmet need for help, did influence the process.

1. Introduction

Major depressive disorder (MDD) and anxiety disorders are highly prevalent and severely disabling, especially when persisting over time (Demyttenaere et al., 2004; Kessler et al., 2015, 2005;Ten Have et al., 2013a,2013b). A substantial number of people with severe symptoms of depression and/or anxiety do not receive adequate treatment (Kohn et al., 2004), with estimates ranging between 30% and 50% (Harvey and Gumport, 2015; Piek et al., 2011; Spijker et al., 2013;

Ten Have et al., 2013a,2013b).

In the Netherlands, most people who suffer from depression and/or anxiety initially go to their general practitioner (GP), who can provide counselling or pharmacotherapy. This can be supplemented with short-term out-patient psychotherapy (5–10 sessions) in primary care,

provided by social workers, social psychiatric nurses, public health nurses or psychologists (Piek et al., 2011; Spijker et al., 2013; van Balkom et al., 2013). In case of severe mental health problems, patients can be referred to specialised mental healthcare. Compared to primary care, specialised treatment is usually characterised by a more multi-disciplinary approach and longer treatment duration. Treatment can consist of individual sessions, group-therapy, day-treatment or in-pa-tient care, and is most often provided by psychiatrists, psychotherapists or psychologists.

According to formal clinical guidelines, referral to specialised mental healthcare can be made by a GP, company doctor or medical specialist. Criteria for referral are an inadequate treatment response to thefirst steps of primary mental healthcare, comorbid disorders, com-plex or severe symptoms such as suicidal behaviour, or chronic or

https://doi.org/10.1016/j.jad.2018.07.072

Received 13 April 2018; Received in revised form 22 June 2018; Accepted 22 July 2018 ⁎_{Corresponding author.}

E-mail address:l.c.kooista@vu.nl(L.C. Kooistra).

Available online 24 July 2018

0165-0327/ © 2018 Elsevier B.V. All rights reserved.

recurring episodes (National Institute for Health and Care Excellence, 2011; Piek et al., 2011; Spijker et al., 2013; van Balkom et al., 2013; van Hemert et al., 2012).

In order to better understand why some patients do not receive adequate care, even though effective treatment is available, it is im-portant to identify the barriers and facilitators for people receiving mental healthcare. This provides healthcare professionals with valuable information on which people are at risk of not getting the services they require and ought to be followed up more thoroughly.

Previous cross-sectional research, examining the characteristics of patients in specialised mental healthcare, showed that patients with more severe or chronic symptoms were more likely to receive some form of specialised mental healthcare (Harris et al., 2015; Piek et al., 2011; ten Have et al., 2013a, 2013b; Verhaak et al., 2009; Wang et al., 2007b, 2007a).

Other factors may also play an important role in the referral process. For example, GP's confidence in their ability to detect mood and anxiety disorders was shown to decrease the probability of referral to specia-lised services, while perceiving more barriers for guideline im-plementation increased the probability (Smolders et al., 2010). Con-cerning patients’ demographic characteristics, several cross-sectional studies found that females, highly educated individuals, people be-tween 30 and 50 years old, and individuals without a paid job more often receive specialised mental healthcare (Alonso et al., 2004; Bijl and Ravelli, 2000; Harris et al., 2015; Kohn et al., 2004; ten Have et al., 2013a, 2013b). Low perceived need for help was found to be an im-portant reason for not receiving treatment, even when the financial consequences of receiving care were minimal and when symptom se-verity was high (Andrews et al., 2001).

Overall, it is still largely unclear to what extent formal clinical Fig. 1. Flow-chart for the selection process of the study cohort.

L.C. Kooistra et al. Journal of Affective Disorders 241 (2018) 226–234

guidelines explain which patients receive specialised mental healthcare, and which other, non-clinical, patient characteristics contribute to this process. The current study expands on previous cross-sectional studies by employing a longitudinal design. The central aim of this study is to examine which depressed or anxious individuals transition to specia-lised mental healthcare during a six-year timeframe. Various potential determinants of transition are examined, including clinical character-istics, demographic charactercharacter-istics, prior treatment history and per-ceived unmet need for help. Examining a broad range of determinants provides the opportunity to establish whether guidelines are followed, and to assess which other factors play a role in the processes of referral and receiving help.

2. Material and methods 2.1. Study sample

Participant data were derived from the Netherlands Study of Depression and Anxiety (NESDA), an on-going longitudinal cohort study investigating the determinants, long-term course and con-sequences of depressive and anxiety disorders (Penninx et al., 2008). At baseline, the total NESDA sample consisted of 2981 adults (18–65 years old) that were included between 2004 and 2007. Follow-up assessments took place after one, two, four, and six years. A detailed overview of the NESDA design and procedures can be found elsewhere (Penninx et al., 2008).

NESDA included participants from different healthcare settings and recruited people in various stages of depressive and/or anxiety dis-orders. Participants were either healthy controls (26%) or fulfilled the DSM-IV-TR criteria (American Psychiatric Association (APA), 2000) for a current or remitted depressive disorder (Major Depressive Disorder or Dysthymia) and/or an anxiety disorder (Social Phobia, Generalised Anxiety Disorder, Agoraphobia, and Panic Disorder with or without Agoraphobia) (74%).

Participants were recruited from the community (19%), primary care (general practice, 54%) and specialised mental healthcare settings (27%), thus representing different healthcare populations. Participants with a primary diagnosis of other severe disorders, such as a psychotic, obsessive-compulsive, bipolar, or severe addiction disorder were ex-cluded from study participation. NESDA was approved by the Ethical Review Board of the VU University Medical Centre and by the local review boards of all participating centres. All participants provided written informed consent.Fig. 1illustrates the selection process of the study cohort for the current study.

The study focused on participants who fulfilled the criteria for a depressive disorder, an anxiety disorder or a comorbid depressive and anxiety disorder in the six months prior to baseline assessment (DSM-IV-TR) (American Psychiatric Association (APA), 2000) who were re-cruited in the general population or in primary care (n = 895). Diag-noses were determined with the Lifetime Composite International Di-agnostic Interview (CIDI) version 2.1 (World Health Organisation (WHO), 1997). As the study aimed to assess long-term determinants of participants transitioning to specialised mental healthcare, participants who already received specialised mental healthcare at baseline were excluded from the study sample. Data on mental healthcare was available for the six months prior to baseline, via the self-report Trimbos/iMTA questionnaire for costs associated with psychiatric illness (TIC-P) (Hakkaart-van Roijen et al., 2002), matching the six-month DSM diagnosis. Excluding participants that reported one or more treatment contacts in specialised mental healthcare prior to baseline led to a sample of n = 759. Fifty-eight participants were ex-cluded from data-analysis because of missing data on all follow-up as-sessments, bringing the total study cohort to n = 701.

2.2. Transition to specialised mental healthcare

The primary outcome variable transition to specialised mental healthcare was defined as three or more visits within one follow-up period to a specialised mental healthcare centre, an independent psy-chiatrist or psychotherapist, and/or a centre specialised in treatment of alcohol or drug abuse or dependence. In order to receive specialised care, patientsfirst require a formal referral by a GP, company doctor or medical professional. Transition can take place either from primary mental healthcare to specialised mental healthcare, or patients can surpass primary care and transition directly to specialised care. The cut-off of three visits was based on the assumption that in Dutch specialised mental healthcare thefirst two sessions are generally focused on the diagnostic phase, with specialised treatment starting from the third session onwards. Information on use of mental healthcare was derived from the TIC-P (Hakkaart-van Roijen et al., 2002), which assessed (mental) healthcare contacts with various healthcare professionals during each follow-up period. The TIC-P was administered one, two, four, and six years after baseline. Time of event (transition to specia-lised mental healthcare) was recorded as the first follow-up year at which participants reported having had three or more specialised mental healthcare contacts since their last assessment.

In order to gain insight into the amount of specialised mental healthcare transitioned participants received during each follow-up period, and the specific setting in which this took place, the number of participants that visited each specialised mental healthcare setting was examined, along with the mean number of contacts in each setting during each follow-up period. At one-year follow-up, only information on the combined number of visits to an independent psychiatrist and/or psychotherapist was available. For two, four, and six-year follow-up, information could be presented separately.

2.3. Determinants of time to transition to specialised mental healthcare 2.3.1. Demographic characteristics

The socio-demographic characteristics gender, age, years of education, total household net income per month (below modal≤ €2400,- and above modal >€2400,-) (Prins et al., 2011a, 2011b; van Beljouw et al., 2010) and employment status (employed, unemployed) were assessed during the baseline interview.

2.3.2. Clinical characteristics

At baseline, the CIDI interview (version 2.1) (World Health Organisation (WHO), 1997) was used to assess whether participants fulfilled the diagnosis of a depressive disorder (Major Depressive Disorder or Dysthymia) and/or an anxiety disorder (Social Phobia, Generalised Anxiety Disorder, Agoraphobia, and Panic Disorder with or without Agoraphobia). In order to examine the predictive value of having a depressive disorder, an anxiety diagnosis, or a combined diagnosis of depression and anxiety, this information was coded into a three-level categorical variable, with depression diagnosis as the reference cate-gory. The CIDI diagnosis Alcohol Dependency was added as a separate determinant to further assess the impact of comorbidity.

In order to gain insight into the predictive value of duration of symptoms, both symptom duration and age of onset of the disorder were assessed. Age of onset of depressive and anxiety disorders was de-termined with the CIDI interview. When participants were diagnosed with comorbid anxiety and/or depressive disorders, the age of onset of the least recent disorder was used. Symptom duration was defined as the percentage of time participants had symptoms of depression, an-xiety or avoidance in the four tofive years prior to baseline. This was assessed with the Life Chart (Lyketsos et al., 1994) which was shown to be an adequate instrument (Denicoff et al., 1997; Honig et al., 2001). When participants reported experiencing symptoms in more than one domain, the domain with the highest percentage was used.

was assessed with the 30-item Inventory of Depressive Symptoms-Self Report (IDS-SR30) (Rush et al., 1996, 1986). In order to assess anxiety severity in the week prior to baseline, the 21-item Beck Anxiety In-ventory (BAI) (Beck et al., 1988) was used. Within the NESDA sample, Cronbach's alpha was 0.93 for both the IDS and BAI questionnaire. Suicidal ideation the week prior to baseline was measured in a semi-structured interview with the five-item Scale for Suicide Ideation (Beck et al., 1979). Participants were considered to have a heightened risk of suicide when they reported suicidal thoughts (a score of at least two out of three) on at least one out of thefive items.

The personality domains agreeableness, neuroticism, conscientious-ness, extraversion and openness to experience were measured with the 60-item NEO personality questionnaire (Costa and McCrae, 1995). Within the NESDA sample Cronbach's alpha ranged from 0.69 (open-ness) to 0.90 (neuroticism).

2.3.3. Prior treatment

The predictive value of prior treatment received was evaluated by assessing whether participants received primary mental healthcare and/or pharmacotherapy prior to baseline. Treatment in primary mental healthcare was measured with the TIC-P (Hakkaart-van Roijen et al., 2002) and was defined as at least three mental healthcare con-tacts in the six months prior to baseline with a GP, psychologist, social worker, and/or social psychiatric nurse. Assessment of pharma-cotherapy was based on medication packages of all psychoactive medication participants used regularly for their depressive or anxious symptoms in the month prior to baseline. The medication was then rated by the interviewer, based on the World Health Organization Anatomical Therapeutic Chemical (ATC) classification system (World Health Organization Collaboration Centre for Drug Statistics Methodology, 2016). It included antidepressants (tricyclic anti-depressants (ATC code N06AA), selective serotonin reuptake inhibitors (ATC code N06AB), other antidepressants (ATC codes N06AF, N06AG, N06AX) and benzodiazepines (ATC codes N03AE, N05BA, N05CD, N05CF).

2.3.4. Perceived need for help

Participants’ perceived unmet need for help in the six months prior to baseline was assessed with the Perceived Need for Care questionnaire (PNCQ), which was found to have an acceptable reliability and validity for this type of study (Meadows et al., 2000). Participants specified whether they received help in the domains of (1) psycho-education and treatment information, (2) medicines and pills, (3) referral to a spe-cialist, (4) psychotherapy and counselling, (5) practical support, and/or (6) skills training during the six months prior to baseline and indicated whether this matched their need. When participants reported that they did not receive help in a domain despite having a need for help, they were considered to have an unmet need for help in this domain. One continuous variable was created, indicating the total amount of do-mains in which participants had a perceived unmet need for help (range 0–6).

2.3.5. Statistical analyses

The probability of transitioning to specialised mental healthcare during follow-up (one, two, four, and six years after baseline) was ex-amined with the Kaplan-Meier estimate. Cox's proportional hazard analyses were used to examine univariate and multivariate associations between time to transition to specialised mental healthcare and possible determinants. The following determinants were examined: (1) partici-pants’ socio-demographic characteristics (age, gender, education level, employment status and household income), (2) clinical factors (CIDI anxiety, depression and/or alcohol dependence diagnoses, age of onset, percentage of time during which participants had symptoms in pastfive years, personality domains, number of chronic somatic diseases under treatment, severity of depressive and anxiety symptoms and suicidal ideation), and (3) participants’ mental healthcare prior to baseline

(psychotropic medication and/or primary mental healthcare contacts) and perceived unmet need for help.

Subjects were right censored in the analysis at the last recorded follow-up or when they did not transition to specialised mental healthcare during the full six-year follow-up period. Time-point of event (transition to specialised mental healthcare) was recorded as the first follow-up at which the event occurred. Time was defined as 1, 2, 4 and 6, reflecting the follow-up time in years since baseline. Proportional hazards were verified in order to rule out interactions between time and the covariate levels. Predictor variables were ex-amined for normality of sampling distribution, univariate outliers and multivariate outliers. Multicollinearity between predictors was checked by calculating the variance inflation factor (VIF; 1/(1-R2_{)) for each} determinant variable in the multivariate model. Values over 2.5 were considered indicative of multicollinearity (Allison, 1999). The analysis was run with the rms function (Harrell, 2018) in R software (R Core Team, 2017).

Continuous variables were standardised, with a unit change re-flecting a decrease or increase of one standard deviation. All Cox ana-lyses were performed using the R survival package (Therneau, 2017). Predictors that were significant at p = .20 (two-tailed) were included in the multivariate model (Mickey and Greenland, 1989). In the multi-variate model a significance level of p < .05 (two-tailed) was applied. In order to examine the combined impact of all significant patient characteristics, a cumulative score was calculated from all determinants that were significantly related to transition to specialised mental healthcare. A logistic regression model was performed with transition as the dependent variable and number of determinants as the in-dependent variable.

3. Results

3.1. Sample characteristics

Mean age of the total study sample was 44.4 years (SD 12.7, range 18–65). The sample consisted of more women than men (n = 506, 72.2%). At baseline, most participants were diagnosed with an anxiety disorder in the past six months (n = 311, 44.4%), 229 participants were diagnosed with both an anxiety and a depressive disorder (32.7%), and the remaining 161 participants (23.0%) were diagnosed with a de-pressive disorder. Table 1 provides an overview of differences in baseline demographic and clinical characteristics between participants who transitioned to specialised mental healthcare and participants who did not.

3.2. Transition to specialised mental healthcare

Over the course of six years, 198 out of 701 participants (28.3%) transitioned to specialised mental healthcare, reporting at least three contacts in a specialised mental healthcare setting. The slope of the survival curve suggests that most participants who transitioned to specialised mental healthcare did so during thefirst year after baseline (n = 66, 33.3%) or in the second year (n = 58, 29.3%). After four years, an additional 35 (17.7%) participants reported a transition to specia-lised mental healthcare, followed by 39 (19.7%) participants at six-year follow-up.Table 2provides an overview per follow-up assessment of the number of transitioned participants that reported receiving treat-ment at each specialised treat-mental healthcare provider (independent psychiatrist or psychotherapist, specialised centre and/or at a centre specialised in treatment of alcohol or drug abuse or dependence) and the mean number of contacts per follow-up.

3.3. Determinants

Univariate Cox regression analyses were used to assess predictive value for time to transition to specialised mental healthcare for (1) the

L.C. Kooistra et al. Journal of Affective Disorders 241 (2018) 226–234

demographic characteristics age, gender, relationship status, years of education, employment status and income, (2) the clinical baseline char-acteristics depressive and/or anxiety disorder, age of onset of current di-agnosis, symptom duration, comorbid alcohol dependence, depression and anxiety severity, suicidal ideation, the NEO personality domains and the number of chronic somatic diseases under treatment, (3) prior pharmaco-logical treatment and primary care, and perceived unmet need for help. Outcomes of univariate analyses can be found inTable 3.

Based on an alpha of 0.20, the baseline determinants age, years of

education, relationship status, age of onset, income level, alcohol depen-dence, depression severity, anxiety severity, suicidal ideation, neuroticism, extraversion, openness to experience, conscientiousness, pharmacological treatment, prior treatment in primary care, and unmet need for help were included in the multivariate model (seeTable 3).

Before running the multivariate model, multicollinearity between predictors was checked by calculating the variance inflation factor (VIF) for each determinant variable in the multivariate model. For de-pression severity a VIF score was found of 2.71, other determinants had a VIF score below cut-off (VIF < 2.5). Based on this finding, the mul-tivariate model was performed both with and without depression se-verity as a determinant. Removing depression sese-verity from the mul-tivariate model did not impact the other outcomes. Therefore, depression severity was kept as a determinant in the multivariate model.

In the multivariate model, age, years of education, suicidal ideation, openness to experience, pharmacological treatment, prior treatment in pri-mary mental healthcare, and perceived unmet need for help significantly predicted time to transition.

Participants who reported suicidal ideation, had a 1.9 times higher probability of transitioning than participants who did not have suicidal ideation (β= 0.657, HR = 1.93 95% CI: 1.34 to 2.78, p < .001). Participants who reported more openness to experience, had a relative probability of 18% per standard deviation (SD = 5.3, β= 0.162, HR = 1.18 95% CI: 1.00 to 1.38, p = .05). Participants who received pharmacological treatment at baseline had 1.45 higher probability of transitioning to specialised care (β= 0.370, HR = 1.45 95% CI: 1.03 to 2.03, p = .033). Participants who received three or more mental healthcare contacts in primary care in the six months prior to baseline, had 1.55 higher probability of transitioning to specialised care (β= 0.437, HR = 1.55 95% CI: 1.13 to 2.12, p = .006). Participants who had an unmet need for help, had a relative probability of 19% per standard deviation (SD = 0.87,β= 0.173, HR = 1.19, 95% CI: 1.06 to 1.34, p = .004). With each standard deviation increase in age the probability of transitioning decreased with 17% (SD = 12.7, β= −0.189, HR = 0.83, 95% CI: 0.70 to 0.98, p = .030). Participants who reported more years of education had a significantly increased probability of transitioning of 27% per standard deviation (SD = 3.3, β= 0.242, HR = 1.27, 95% CI: 1.09 to 1.49, p = .003). The predictive value of the set of determinants was R2= 0.14 (estimated 95% CI: 0.08 to 0.18, using R-square software (R Core team, 2017)).

Table 4 presents the odds of transitioning to specialised mental healthcare for subject groups with various risk profiles. In order to calculate the cumulative score, continuous variables were dichotomised based on median split: age (≥ 48 years old), years of education (≥ 11), openness of experience (≥ 26). For perceived unmet need for help, a cut-off of one domain or more was used.

Because only 39 participants did not have any significant determi-nants in their profile, participants with zero to one determidetermi-nants were grouped together, and used as a reference group (n = 144). Compared to the reference group, the odds of transition did not significantly in-crease when participants had two determinants in their profile Table 1

Baseline characteristics of the study sample.

Baseline characteristics No transition Transition to SMHC Total N = 503 N = 198 N = 701 Demographics Age (mean ± SD) 45.3 (12.8) 42.1 (12.3) 44.4 (12.7) Gender: Females (n, %) 367 (73.0) 139 (70.2) 506 (72.2) In a relationship (yes, n, %) 352 (70.0) 125 (63.1) 477 (68.0) Years of education (mean ± SD) 11.4 (3.2) 12.6 (3.5) 11.7 (3.3) Employed (n, %) 424 (84.3) 163 (82.3) 587 (83.7) Household income per month (n,

%) Below modal (≤ €2400,-) 316 (45.1) 187 (26.7) 450 (64.2) Above modal (>€2400,-) 134 (19.1) 64 (9.1) 251 (35.8) Clinical characteristics Six-month diagnosis (n, %) Depression 111 (22.1) 50 (25.3) 161 (23.0) Anxiety 232 (46.1) 79 (39.9) 311 (44.4) Comorbid Depression/Anxiety 160 (31.8) 69 (34.8) 229 (32.7) Age of onset (mean ± SD) 22.9 (13.6) 19.2 (11.4) 21.9 (13.1) Symptom duration (mean ± SD) 45.0 (35.3) 46.6 (33.5) 45.5 (34.8) Diagnosis alcohol abuse/

dependence (n, %)

146 (29.0) 78 (39.4) 224 (32.0) Depression severity

(mean ± SD)

24.5 (10.9) 28.6 (10.7) 25.7 (11.0) Anxiety severity (mean ± SD) 14.2 (9.1) 16.1 (9.7) 14.7 (9.3) Suicidal ideation in past week (n,

%) 47 (9.3) 50 (25.3) 97 (13.8) Personality characteristics (mean ± SD) Neuroticism 39.2 (7.3) 41.1 (6.5) 39.7 (7.1) Extraversion 35.6 (6.7) 34.5 (6.5) 35.3 (6.6) Agreeableness 43.1 (5.2) 42.9 (5.2) 43.0 (5.2) Openness 30.7 (5.3) 32.5 (5.2) 31.2 (5.4) Conscientiousness 37.1 (5.7) 35.9 (6.4) 36.8 (5.9) # of somatic diseases (mean ± SD) No somatic disease 246 (48.9) 96 (48.5) 342 (48.8) One somatic disease 167 (33.2) 67 (33.8) 234 (33.4) Two or more somatic diseases 90 (17.9) 35 (17.7) 125 (17.8) Psychotropic medication (n, %) 118 (23.5) 57 (28.8) 175 (25.0) Primary mental healthcare (n, %)

Three or more contacts 144 (28.6) 90 (45.5) 234 (33.4) Unmet need for help

(mean ± SD)

0.30 (0.78) 0.65 (1.00) 0.40 (0.87)

Note. SD: Standard Deviation; SMHC: Specialised Mental Healthcare.

Table 2

Overview of specialised mental healthcare received by transitioned participants: mean number of visits per follow-up.

Specialised centre Psychiatrist2 _{Psychotherapist}2 _{Centre for alcohol/drugs}

Year N1 _{n Contacts n Contacts n Contacts n Contacts}

Mean (SD) Mean (SD) Mean (SD) Mean (SD) One 66 24 13.7 (22.1) 39 11.3 (5.5) – – 3 8.0 (4.9) Two 58 36 20.2 (23.8) 7 12.4 (8.5) 13 13.4 (12.1) 2 11.5 (5.0) Four 35 17 35.2 (34.9) 7 15.4 (18.4) 12 12.1 (9.3) 3 33.7 (29.7) Six 39 23 16.9 (26.3) 7 17.4 (25.6) 11 23.0 (30.1) 2 16.0 (5.7)

(n = 204, OR = 1.61, 95% CI: 0.92–2.89, p = .102). The odds did in-crease significantly when participants had three (n = 187, OR = 2.44, 95% CI: 1.44–4.35, p = .002), or four or more determinants (OR = 5.25, 95% CI: 3.06–9.32, p < .001). This finding suggests there is a cumulative effect of the number of determinants present in patients’ profiles, increasing the probability that participants transitioned to specialised mental healthcare.

4. Discussion

This longitudinal study examined possible determinants of partici-pants with a current DSM-IV diagnosis of depression and/or anxiety transitioning to specialised mental healthcare over the course of six

years. Out of 701 participants, 198 (28.3%) transitioned to specialised mental healthcare. Transition most often took place within two years after baseline (n = 124, 63%). In a multivariate model, younger age, years of education, suicidal ideation, openness to experience, prior pharmacological treatment and primary mental healthcare, and a per-ceived unmet need for help, significantly predicted time to transition. When participants' profiles included three or more determinants, the probability of transitioning to specialised care increased significantly.

4.1. Clinical profile and need for help

Participants who already received pharmacotherapy or primary mental healthcare at baseline, had an increased probability of transi-tioning to specialised services compared to participants who were not yet receiving one of these types of care. Thisfinding seems to be in line with treatment and referral guidelines, which focus on a stepped-care approach and encourage healthcare professionals to refer patients who still experience symptoms after the first steps of mental healthcare (National Institute for Health and Care Excellence, 2011; Piek et al., 2011; Spijker et al., 2013; van Balkom et al., 2013). However, this explanation might be less relevant for participants who did not transi-tion until four to six years after the baseline measurement. In additransi-tion, prior treatment could have been aimed at a diagnosis or problem other than depression or anxiety.

When participants reported that they did not receive help despite having a need for help (perceived unmet need for help), they had an increased probability of transitioning to specialised care. This is in line with prior research (Andrews et al., 2001). Possibly, participants who Table 3

Hazard ratio's: relative probability for transition to specialised mental healthcare.

Univariate analyses Multivariate analyses (n = 689)

N HR 95% CI p-value HR 95% CI p-value Demographics

Age (mean ± SD)* 701 0.792 0.69; 0.91 .001 0.827 0.70; 0.98 .030 Gender: Females (n, %) 701 0.921 0.68; 1.25 .599

In a relationship (yes, n, %) 701 0.783 0.59; 1.05 .097 0.903 0.65; 1.25 .538 Years of education (mean ± SD)* 701 1.319 1.15; 1.51 <.001 1.270 1.09; 1.49 .003 Employed (n, %) 701 0.886 0.62; 1.28 .514

Household income per month (n, %) 701

Below modal (≤ €2400,-) Ref.

Above Modal (>€2400,-) 0.819 0.61; 1.10 .185 0.857 0.72; 1.03 .379 Clinical characteristics Six-month diagnosis (n, %) 701 Depression Ref. Anxiety 0.808 0.57; 1.15 .238 Comorbid Depression/Anxiety 1.004 0.70; 1.45 .982

Age of onset (mean ± SD)* 697 0.777 0.67; 0.90 .001 0.863 0.61; 1.21 .099 Symptom duration (mean ± SD)* 694 1.030 0.90; 1.18 .668

Diagnosis alcohol abuse/dependence (n, %) 701 1.460 1.10; 1.94 .009 1.185 0.88; 1.60 .266 Depression severity (mean ± SD)* 696 1.530 1.29; 1.82 <.001 1.329 0.99; 1.78 .055 Anxiety severity (mean ± SD)* 697 1.247 1.07; 1.45 .004 0.913 0.73; 1.15 .435 Suicidal ideation in past week (n, %) 701 2.616 1.90; 3.61 <.001 1.929 1.34; 2.78 <.001 Personality characteristics (mean ± SD)*

Neuroticism 696 1.401 1.16; 1.70 <.001 0.971 0.74; 1.27 .830 Extraversion 696 0.854 0.73; 1.00 .048 0.985 1.00; 1.38 .883 Agreeableness 696 0.941 0.82; 1.09 .409

Openness 696 1.305 1.14; 1.50 <.001 1.178 0.80; 1.21 .046 Conscientiousness 696 0.819 0.71; 0.95 .007 0.953 0.81; 1.13 .569 # of somatic diseases (mean ± SD) 701

No somatic disease Ref

One somatic disease 1.024 0.75; 1.40 .880 Two or more somatic diseases 1.004 0.68; 1.48 .983

Psychotropic medication (n, %) 701 1.239 0.91; 1.69 .172 1.447 1.03; 2.03 .033 Primary mental healthcare (n, %)

Three or more contacts 701 1.902 1.44; 2.52 <.001 1.548 1.13; 2.12 .006 Unmet need for help (mean ± SD)* 701 1.295 1.18; 1.43 <.001 1.189 1.06; 1.34 .004

Note. * Standardised for Cox regression analyses; SD: Standard Deviation; SMHC: Specialised Mental Healthcare. In bold: p-value significant (univariate: a > 0.20, multivariate: a > 0.05).

Table 4

Number of determinants present in participants per group and odds of transi-tioning to specialised mental healthcare.

# of determinants No SMHC (n = 503) SMHC (n = 198) Total sample (n = 701) Odds ratio (95% CI) n (%) n (%) n (%) 0–1 123 (24.5) 21 (10.6) 144 (20.6) Ref 2 160 (31.8) 44 (22.2) 204 (29.1) 1.61 (0.92; 2.89) 3 132 (26.2) 55 (27.8) 187 (26.7) 2.44 (1.41; 4.35)** 4+ 88 (17.5) 78 (39.4) 166 (23.7) 5.25 (3.06; 9.32)***

Note. SMHC: specialised mental healthcare. ** p < .01; *** p < .001.

L.C. Kooistra et al. Journal of Affective Disorders 241 (2018) 226–234

were aware of their need for help were more inclined to start specia-lised treatment than participants who were not. Another explanation is that healthcare professionals take their patients’ needs into account when determining mental healthcare trajectories. However, whether this was actually the case could not be examined in the current study. In the current study, comorbid disorders, symptom duration and self-reported symptom severity lost their univariate predictive value within the multivariate model. This was unexpected, as these variables are also important referral criteria and prior cross-sectional research did suggest that individuals with more severe symptoms are more likely to receive mental healthcare (Ten Have et al., 2013a,2013b; Verhaak et al., 2009; Wang et al., 2007a, 2007b). Part of thisfinding can be explained by the predictive value of suicidal ideation, as this is also a severity marker. In the NESDA sample for example, suicidal partici-pants were shown to be more severely ill than participartici-pants with de-pression or anxiety who did not display suicidal ideation (Stringer et al., 2013). However, suicidal ideation was reported by a quarter of tran-sitioned participants, indicating that three quarters of participants transitioned to specialised mental healthcare based on other criteria. 4.2. Demographic characteristics

The predictive value of age matched earlier studies (Alonso et al., 2004; Kohn et al., 2004), suggesting that persons who are older have a lower probability of transitioning to specialised mental healthcare than younger persons with a similar clinical profile. Even though age is not included in formal guidelines, it appears to be worthwhile for GPs to more closely monitor this older patient group.

In the current study, years of education also predicted transition to specialised mental healthcare independently of clinical severity char-acteristics, while controlling for other demographic variables such as employment and household income. More highly educated participants had a higher probability of transitioning to specialised services. Possibly, more highly educated individuals are better at voicing their needs and explaining their symptoms, thereby increasing their prob-ability of referral.

4.3. Strengths and limitations

This study was, to our knowledge, one of the first to employ a longitudinal epidemiological design in order to assess the long-term mental healthcare trajectories of people with an anxiety and/or de-pressive disorder. Examining a broad range of determinants provides opportunity to examine which factors play a role in the process of re-ferral and help seeking.

The current set of determinants does pose some methodological challenges, such as the determinant to event ratio, multiple testing and (multi)collinearity between determinants. To examine the precise in-dividual contribution of determinants to the joint prediction, a larger sample would be required. As the number of events to determinant ratio decreases, the estimations become more biased. However, the de-terminant to event ratio (1:9) was close to the threshold of 1:10 that was suggested by Peduzzi and colleagues for proportional Hazard re-gression analyses (Peduzzi et al., 1995). Further, while collinearity was present, for all but one determinant (depression severity) the values were within the acceptable margin. Removing depression severity as a determinant from the multivariate model did not affect outcomes.

There are also some considerations, regarding sample selection and outcome measures, to be taken into account when interpreting the data. Future studies are advised to examine the extent to which participants experience stigma, because this could also have a major impact on mental health, identification of mental disorders and healthcare tra-jectories (Prins et al., 2011a, 2011b; Whiteford et al., 2013).

Participants and their mental healthcare professionals were not necessarily aware or in agreement that a mental disorder was present. This likely will have impacted transition to specialised care. In the

current study this information could not be taken into account, because it was not available for all participants. A previous NESDA study ex-amining patient records in the primary care sample, showed that GP's correctly recognised 69% of the depressed participants and 81% of non-depressed participants (Joling et al., 2011). This suggests that while identification rates by GP's were high, a substantial number of patients might have been under-diagnosed or over-diagnosed.

Participants were included when they did not receive specialised mental healthcare in the months prior to baseline, matching a six-month DSM diagnosis. No criterion could be added for specialised treatment before this time-period. Consequently, it is possible that, for a subgroup of participants, a return to specialised services was measured, rather than afirst transition. Possibly, clinical profiles in this group differed from those of participants who did not receive specialised treatment before, for example in terms of symptom severity or duration. It would be interesting for future studies to examine the full mental healthcare trajectories over time, rather than to focus on one transition. Combined with specific information on type of treatment received, treatment response and remission rates, this could add to the knowl-edge on treatment availability and effectiveness. Also, an interesting next step in future research would be to examine the reasons for referral and transition to specialised mental healthcare according to both pa-tients and healthcare professionals. In this context, a mixed-method approach could be considered. This could shed more light on barriers and facilitators for receiving mental healthcare.

5. Conclusions

Combining all information, the current study suggests that transi-tion to specialised mental healthcare could only be partly explained by clinical characteristics that are included in official clinical referral guidelines.

The clinical characteristics suicidal ideation and prior treatment proved to be important determinants. However, other clinical factors such as comorbidity, symptom duration and symptom severity did not contribute to the probability of transition to specialised mental healthcare. Age, education level, openness to experience and perceived unmet need for help were related to transition to specialised services, even though these factors are not included in guidelines. Participants with combined profiles including three or more determinants, were shown to have an increased probability of transitioning to specialised mental healthcare. Groups that might be at risk of under-treatment according to thefindings in this study are older adults and people with lower education. It would therefore be advisable for mental healthcare professionals to pay specific attention to these patients, making sure that they receive the services they require.

Declaration of interest None.

Authors’ contributions

BP is principal investigator of NESDA. LK, JW and JR collaborated in the design of this particular study. BP, HR and PvO critically revised the design. LK performed the statistical analyses and wrote the manu-script. JR supervised data analysis and the interpretation of the data. JW supervised writing of this manuscript. All authors commented on several drafts, as well as read and approved thefinal manuscript. Author statement

Conflict of interest

The authors declare that there are no conflicts of interest. Role of funding source

The funding source had no role in the design of this study, its ex-ecution, analyses, interpretation of the data, or decision to submit re-sults.

Acknowledgements

The infrastructure for the NESDA study (www.nesda.nl) is funded through the Geestkracht program of the Netherlands Organisation for Health Research and Development (ZonMw, grant number 10-000-1002) and financial contributions by participating universities and mental healthcare organizations (VU University Medical Center, GGZ inGeest, Leiden University Medical Center, Leiden University, GGZ Rivierduinen, University Medical Center Groningen, University of Groningen, Lentis, GGZ Friesland, GGZ Drenthe, Rob Giel Onderzoekscentrum).

References

Allison, P.D., Logistic regression using the SAS system: theory and application, 1999, SAS Institute; Cary, NC.

Alonso, J., Angermeyer, M.C., Bernert, S., Bruffaerts, R., Brugha, T.S., Bryson, H., de Girolamo, G., Graaf, R., Demyttenaere, K., Gasquet, I., Haro, J.M., Katz, S.J., Kessler, R.C., Kovess, V., Lépine, J.P., Ormel, J., Polidori, G., Russo, L.J., Vilagut, G., Almansa, J., Arbabzadeh-Bouchez, S., Autonell, J., Bernal, M., Buist-Bouwman, M.A., Codony, M., Domingo-Salvany, A., Ferrer, M., Joo, S.S., Martínez-Alonso, M., Matschinger, H., Mazzi, F., Morgan, Z., Morosini, P., Palacín, C., Romera, B., Taub, N., Vollebergh, W.A.M., 2004. Use of mental health services in Europe: Results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr. Scand. Suppl. 109, 47–54.https://doi.org/10.1111/j.1600-0047.2004.00330.x.

American Psychiatric Association (APA), 2000. Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition). American Psychiatric Association, Washington, DC.

Andrews, G., Henderson, S., Hall, W., 2001. Prevalence, comorbidity, disability and service utilisation. Overview of the Australian national mental health survey. Br. J. Psychiatry 178, 145–153.https://doi.org/10.1192/bjp.178.2.145.

Beck, A.T., Epstein, N., Brown, G., Steer, R.A., 1988. An inventory for measuring clinical anxiety: psychometric properties. J. Consult. Clin. Psychol. 56, 893–897.https://doi. org/10.1037/0022-006X.56.6.893.

Beck, A.T., Kovacs, M., Weissman, A., 1979. Assessment of suicidal intention: The scale for suicide ideation. J. Consult. Clin. Psychol. 47, 343–352.https://doi.org/10.1037/ 0022-006X.47.2.343.

Bijl, R., Ravelli, A., 2000. Psychiatric morbidity, service use, and need for care in the general population: results of the Netherlands mental health survey and incidence study. Am. J. Public Health 90, 602–607.https://doi.org/10.2105/ajph.90.4.602. Costa, P., McCrae, R.R., 1995. Inventory domains and facets: Hierarchical personality assessment using the revised NEO Personality Inventory. J. Pers. Assess. 64, 21–50.

https://doi.org/10.1207/s15327752jpa6401.

Demyttenaere, K., Bruffaerts, R., Posada-Villa, J., Gasquet, I., Kovess, V., Lepine, J.P., Angermeyer, M.C., Bernert, S., de Girolamo, G., Morosini, P., Polidori, G., Kikkawa, T., Kawakami, N., Ono, Y., Takeshima, T., Uda, H., Karam, E.G., Fayyad, J.A., Karam, A.N., Mneimneh, Z.N., Medina-Mora, M.E., Borges, G., Lara, C., de Graaf, R., Ormel, J., Gureje, O., Shen, Y., Huang, Y., Zhang, M., Alonso, J., Haro, J.M., Vilagut, G., Bromet, E.J., Gluzman, S., Webb, C., Kessler, R.C., Merikangas, K.R., Anthony, J.C., Von Korff, M.R., Wang, P.S., Brugha, T.S., Aguilar-Gaxiola, S., Lee, S., Heeringa, S., Pennell, B.-E., Zaslavsky, A.M., Ustun, T.B., Chatterji, S., WHO World Mental Health Survey Consortium, 2004. Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys. JAMA 291, 2581–2590.https://doi.org/10.1001/jama.291.21.2581.

Denicoff, K.D., Smith-Jackson, E.E., Disney, E.R., Suddath, R.L., Leverich, G.S., Post, R.M., 1997. Preliminary evidence of the reliability and validity of the prospective life-chart methodology (LCM-p). J. Psychiatr. Res. 31, 593–603.https://doi.org/10.1016/ S0022-3956(96)00027-1.

Hakkaart-van Roijen, L., van Straten, A., Tiemens, B., Donker, M.C.H., 2002. Trimbos/ iMTA Questionnaire For Costs Associated With Psychiatric Illness (TIC-P). Institute of Medical Technology Assessment (iMTA), Rotterdam.

Harrell, F.E., Regression Modeling Strategies, 2018, R package version 5.1-2. Harris, M.G., Baxter, A.J., Reavley, N., Diminic, S., Pirkis, J., Whiteford, H.A., 2015.

Gender-related patterns and determinants of recent help-seeking for past-year af-fective, anxiety and substance use disorders: Findings from a national epidemiolo-gical survey. Epidemiol. Psychiatr. Sci. 1–14.https://doi.org/10.1017/ S2045796015000876.

Harvey, A.G., Gumport, N.B., 2015. Evidence-based psychological treatments for mental

disorders: modifiable barriers to access and possible solutions. Behav. Res. Ther. 68, 1–12.https://doi.org/10.1016/j.brat.2015.02.004.

Honig, A., Hendriks, C.H., Akkerhuis, G.W., Nolen, W.A., 2001. Usefulness of the retro-spective Life-Chart method manual in outpatients with a mood disorder: A feasibility study. Patient Educ. Couns. 43, 43–48.https://doi.org/10.1016/S0738-3991(00) 00144-0.

Joling, K.J., van Marwijk, H.W.J., Piek, E., van der Horst, H.E., Penninx, B.W.J.H., Verhaak, P., van Hout, H.P.J., 2011. Do GPs’ medical records demonstrate a good recognition of depression? A new perspective on case extraction. J. Affect. Disord. 133, 522–527.https://doi.org/10.1016/j.jad.2011.05.001.

Kessler, R.C., Chiu, W.T., Demler, O., Merikangas, K.R., Walters, E.E., 2005. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the national comorbidity survey replication. Arch. Gen. Psychiatry 62, 617–627.https://doi.org/10.1001/ archpsyc.62.6.617.

Kessler, R.C., Sampson, N.A., Berglund, P., Gruber, M.J., Al-Hamzawi, A., Andrade, L., Bunting, B., Demyttenaere, K., Florescu, S., de Girolamo, G., Gureje, O., He, Y., Hu, C., Huang, Y., Karam, E., Kovess-Masfety, V., Lee, S., Levinson, D., Medina Mora, M.E., Moskalewicz, J., Nakamura, Y., Navarro-Mateu, F., Browne, M.A.O., Piazza, M., Posada-Villa, J., Slade, T., ten Have, M., Torres, Y., Vilagut, G., Xavier, M., Zarkov, Z., Shahly, V., Wilcox, M.A., 2015. Anxious and non-anxious major depressive disorder in the World Health Organization World Mental Health Surveys. Epidemiol. Psychiatr. Sci. 24, 210–226.https://doi.org/10.1017/S2045796015000189. Kohn, R., Saxena, S., Levav, I., Saraceno, B., 2004. The treatment gap in mental health

care. Bull. World Health Organ. 82, 858–866. https://doi.org/10.1590/S0042-96862004001100011.

Lyketsos, C.G., Nestadt, G., Cwi, J., Heithoff, K., 1994. The life chart interview: A stan-dardized method to describe the course of psychopathology. Int. J. Methods Psychiatr. Res. 4, 143–155.

Meadows, G., Harvey, C., Fossey, E., Burgess, P., 2000. Assessing perceived need for mental health care in a community survey: Development of the Perceived Need for Care Questionnaire (PNCQ). Soc. Psychiatry Psychiatr. Epidemiol. 35, 427–435.

Mickey, R.M., Greenland, S., 1989. The impact of confounder selection criteria on effect estimation. Am. J. Epidemiol. 129, 125–137.

National Institute for Health and Care Excellence, Common Mental Health Problems : Identification and Pathways to Care, 2011, National Institute for Health and Care Excellence, Manchester, England.

Peduzzi, P., Concato, J., Feinstein, A.R., Holford, T.R., 1995. Importance of events per independent variable in proportional hazards regression analysis II. Accuracy and precision of regression estimates. J. Clin. Epidemiol. 48, 1503–1510.https://doi.org/ 10.1016/0895-4356(95)00048-8.

Penninx, B.W.J.H., Beekman, A.T.F., Smit, J.H., Zitman, F.G., Nolen, W.A., Spinhoven, P., Cuijpers, P., de Jong, P.J., van Marwijk, H.W.J., Assendelft, W.J.J., van der Meer, K., Verhaak, P., Wensing, M., de Graaf, R., Hoogendijk, W.J., Ormel, J., van Dyck, R., NESDA Research Consortium, 2008. The Netherlands Study of Depression and Anxiety (NESDA): Rationale, objectives and methods. Int. J. Methods Psychiatr. Res. 17, 121–140.https://doi.org/10.1002/mpr.256.

Piek, E., van der Meer, K., Penninx, B.W.J.H., Verhaak, P.F.M., Nolen, W.a, 2011. Referral of patients with depression to mental health care by Dutch general practitioners: An observational study. BMC Fam. Pract. 12, 41. https://doi.org/10.1186/1471-2296-12-41.

Prins, M., Meadows, G., Bobevski, I., Graham, A., Verhaak, P., Van derMeer, K., Penninx, B.W.J.H., Bensing, J., 2011a. Perceived need for mental health care and barriers to care in the Netherlands and Australia. Soc. Psychiatry Psychiatr. Epidemiol. 46, 1033–1044.https://doi.org/10.1007/s00127-010-0266-3.

Prins, M.A., Verhaak, P.F., Hilbink-Smolders, M., Spreeuwenberg, P., Laurant, M.G., van der Meer, K., van Marwijk, H.W., Penninx, B.W., Bensing, J.M., 2011b. Outcomes for depression and anxiety in primary care and details of treatment: a naturalistic longitudinal study. BMC Psychiatry 11, 180. https://doi.org/10.1186/1471-244X-11-180.

R Core Team, 2017. R: A Language and Environment For Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

Rush, A.J., Giles, D.E., Schlesser, M.A., Fulton, C.L., Weissenburger, J., Burns, C., 1986. The Inventory for Depressive Symptomatology (IDS): Preliminaryfindings. Psychiatry Res 18, 65–87.https://doi.org/10.1016/0165-1781(86)90060-0.

Rush, A.J., Gullion, C.M., Basco, M.R., Jarrett, R.B., Trivedi, M.H., 1996. The Inventory of Depressive Symptomatology (IDS): psychometric properties. Psychol. Med. 26, 477–486.

Smolders, M., Laurant, M., Verhaak, P., Prins, M., van Marwijk, H., Penninx, B.W.J.H., Wensing, M., Grol, R., 2010. Which physician and practice characteristics are asso-ciated with adherence to evidence-based guidelines for depressive and anxiety dis-orders? Med. Care 48, 240–248.https://doi.org/10.1097/MLR.0b013e3181ca27f6.

Spijker, J., Bockting, C.L.H., Meeuwissen, J.A.C., van Vliet, I.M., Emmelkamp, P.M.G., Hermens, M.L.M., van Balkom, A.J.L.M., 2013. Multidisciplinaire Richtlijn Depressie (Derde Revisie). Richtlijn voor de Diagnostiek, Behandeling en Begeleiding van Volwassen Patienten met een Depressieve Stoornis. Trimbos-instituut, Utrecht, the Netherlands [Multidisciplinary guideline for depression (third revision)]. Stringer, B., van Meijel, B., Eikelenboom, M., Koekkoek, B., Verhaak, P.F.M., Kerkhof,

A.J.M.F., Penninx, B.W.J.H., Beekman, A.T.F., 2013. Perceived need for care and health care utilization among depressed and anxious patients with and without sui-cidal ideation. Crisis 34, 192–199.https://doi.org/10.1027/0227-5910/a000182. Ten Have, M., de Graaf, R., van Dorsselaer, S., Beekman, A., 2013. Lifetime treatment

contact and delay in treatment seeking afterfirst onset of a mental disorder. Psychiatr. Serv. 64, 981–989.https://doi.org/10.1176/appi.ps.201200454. Ten Have, M., Nuyen, J., Beekman, A., de Graaf, R., 2013. Common mental disorder

severity and its association with treatment contact and treatment intensity for mental health problems. Psychol. Med. 43, 2203–2213.https://doi.org/10.1017/

L.C. Kooistra et al. Journal of Affective Disorders 241 (2018) 226–234

S0033291713000135.

Therneau, T.M., Survival Analysis, 2017, R package version 2.42-3.

Van Balkom, A.J.L.M., van Vliet, I.M., Emmelkamp, P.M.G., Bockting, C.L.H., Spijker, J., Hermens, M.L.M., Meeuwissen, J.A.C., 2013. Multidisciplinaire Richtlijn Angststoornissen (derde revisie). Richtlijn voor de Diagnostiek, Behandeling en Begeleiding van Volwassen Patiënten met een Angststoornis. Trimbos-instituut, Utrecht, the Netherlands [Multidisciplinary guideline for anxiety disorders (third revision)].

Van Beljouw, I.M.J., Verhaak, P.F.M., Cuijpers, P., van Marwijk, H.W.J., Penninx, B.W.J.H., 2010. The course of untreated anxiety and depression, and determinants of poor one-year outcome: a one-year cohort study. BMC Psychiatry 10, 86.https://doi. org/10.1186/1471-244x-10-86.

Van Hemert, A.M., Kerkhof, A.J.F.M., De Keijser, J., Verwey, B., van Boven, C., Hummelen, J.W., De Groot, M.H., Lucassen, P., Meerdinkveldboom, J., Steendam, M., Stringer, B., Verlinde, A.A., van de Glind, G., 2012. Multidisciplinaire richtlijn Diagnostiek en Behandeling van Suïcidaal Gedrag. Trimbos-instituut, Utrecht, the Netherlands [Multidisciplinary guideline for suicidal behaviour].

Verhaak, P.F.M., Prins, M.A., Spreeuwenberg, P., Draisma, S., van Balkom, A.J.L.M., Bensing, J.M., Laurant, M.G.H., van Marwijk, H.W.J., van der Meer, K., Penninx, B.W.J.H., 2009. Receiving treatment for common mental disorders. Gen. Hosp. Psychiatry 31, 46–55.https://doi.org/10.1016/j.genhosppsych.2008.09.011. Wang, P.S., Aguilar-Gaxiola, S., Alonso, J., Angermeyer, M.C., Borges, G., Bromet, E.J.,

Bruffaerts, R., de Girolamo, G., de Graaf, R., Gureje, O., Haro, J.M., Karam, E.G., Kessler, R.C., Kovess, V., Lane, M.C., Lee, S., Levinson, D., Ono, Y., Petukhova, M.,

Posada-Villa, J., Seedat, S., Wells, J.E., 2007a. Worldwide use of mental health ser-vices for anxiety, mood, and substance disorders: Results from 17 countries in the WHO world mental health surveys. Lancet 370, 841–850.https://doi.org/10.1016/ S0140-6736(07)61414-7.Worldwide.

Wang, P.S., Angermeyer, M., Borges, G., Bruffaerts, R., Tat Chiu, W., de Girolamo, G., Fayyad, J., Gureje, O., Haro, J.M., Huang, Y., Kessler, R.C., Kovess, V., Levinson, D., Nakane, Y., Oakley Brown, M.A., Ormel, J.H., Posada-Villa, J., Aguilar-Gaxiola, S., Alonso, J., Lee, S., Heeringa, S., Pennell, B.-E., Chatterji, S., Ustün, T.B., 2007b. Delay and failure in treatment seeking afterfirst onset of mental disorders in the World Health Organization's world mental health survey initiative. World Psychiatry 6, 177–185.

Whiteford, H.A., Degenhardt, L., Rehm, J., Baxter, A.J., Ferrari, A.J., Erskine, H.E., Charlson, F.J., Norman, R.E., Flaxman, A.D., Johns, N., Burstein, R., Murray, C.J.L., Vos, T., 2013. Global burden of disease attributable to mental and substance use disorders: Findings from the Global Burden of Disease Study 2010. Lancet 382, 1575–1586.https://doi.org/10.1016/S0140-6736(13)61611-6.

World Health Organisation (WHO), 1997. Composite International Diagnostic Interview, Core version 2.1: Interviewer's Manual. World Health Organisation, Geneva, Switzerland.